WO2021163986A1 - Procédé et système de planification de fabrication de semi-conducteurs - Google Patents

Procédé et système de planification de fabrication de semi-conducteurs Download PDF

Info

Publication number
WO2021163986A1
WO2021163986A1 PCT/CN2020/076111 CN2020076111W WO2021163986A1 WO 2021163986 A1 WO2021163986 A1 WO 2021163986A1 CN 2020076111 W CN2020076111 W CN 2020076111W WO 2021163986 A1 WO2021163986 A1 WO 2021163986A1
Authority
WO
WIPO (PCT)
Prior art keywords
load
fabrication
kpis
schedule
current
Prior art date
Application number
PCT/CN2020/076111
Other languages
English (en)
Inventor
Weirong ZHENG
Heqing WANG
Original Assignee
Yangtze Memory Technologies Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Yangtze Memory Technologies Co., Ltd. filed Critical Yangtze Memory Technologies Co., Ltd.
Priority to PCT/CN2020/076111 priority Critical patent/WO2021163986A1/fr
Priority to CN202080000366.2A priority patent/CN111356988B/zh
Priority to TW109110944A priority patent/TWI788650B/zh
Priority to US16/905,194 priority patent/US11385627B2/en
Publication of WO2021163986A1 publication Critical patent/WO2021163986A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
    • G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
    • G05B13/027—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion using neural networks only
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5083—Techniques for rebalancing the load in a distributed system
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06316—Sequencing of tasks or work
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/67005—Apparatus not specifically provided for elsewhere
    • H01L21/67011—Apparatus for manufacture or treatment
    • H01L21/67155—Apparatus for manufacturing or treating in a plurality of work-stations
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B17/00—Systems involving the use of models or simulators of said systems
    • G05B17/02—Systems involving the use of models or simulators of said systems electric
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32301—Simulate production, process stages, determine optimum scheduling rules
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32333—Use of genetic algorithm
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32335—Use of ann, neural network
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/34—Director, elements to supervisory
    • G05B2219/34379—Job management
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P80/00—Climate change mitigation technologies for sector-wide applications
    • Y02P80/40—Minimising material used in manufacturing processes

Definitions

  • This application relates to the field of semiconductor manufacturing and, specifically, to a semiconductor fabrication scheduling method and system.
  • a state-of-art integrated circuit (IC) chip often requires hundreds or even thousands process steps performed on a single semiconductor wafer via different and same types of process machines.
  • the lots In order to provide fabrication of product lots from a manufacturing line within an enterprise to meet committed schedules, the lots must be scheduled to dispatch to each tool or piece of manufacturing equipment expeditiously. Further, to avoid having either long waiting or queuing times and having pieces of manufacturing equipment that are “starving” or have no queue, the scheduling must account for both the amount of product to be immediately processed and the capacity, capability, and present queue length of pieces of manufacturing equipment following the current piece of manufacturing equipment.
  • the scheduling problem i.e., how product lots should be released into the fabrication process and dispatched among pieces of manufacturing equipment for processing, is among major fabrication control problems facing the manufacturers.
  • the disclosed methods and systems are directed to solve one or more problems set forth above and other problems.
  • One aspect of the present disclosure includes a semiconductor fabrication scheduling method.
  • the method includes creating a load scheduling data schema including facility data of product lots to be dispatched to a plurality of workstations; and generating a load schedule profile using a load-balancing model and based on the load scheduling data schema.
  • the load-balancing model includes one or more objective functions and there is at least one weight factor in an objective function.
  • the method also includes generating a current load schedule based on the load schedule profile; dispatching the product lots to the plurality of workstations using the current load schedule to complete fabrication of the product lots; obtaining a set of current key performance indicators (KPIs) of the completed fabrication of the product lots; and automatically adjusting the weight factors of the objective functions of the load-balancing model based on the current KPIs using a big-data architecture to generate a next load schedule for next cycle of fabrication.
  • KPIs current key performance indicators
  • the system includes a plurality of workstations; a server; and a dispatcher controlled by the server and dispatching product lots to be fabricated to the plurality of workstations.
  • the server is configured to perform: creating a load scheduling data schema including facility data of the product lots; generating a load schedule profile using a load-balancing model and based on the load scheduling data schema, wherein the load-balancing model includes one or more objective functions and there is at least one weight factor in an objective function; generating a current load schedule based on the load schedule profile; controlling the dispatcher to dispatch the product lots to the plurality of workstations using the current load schedule to complete fabrication of the product lots; obtaining a set of current key performance indicators (KPIs) of the completed fabrication of the product lots; and automatically adjusting the weight factors of the objective functions of the load-balancing model based on the current KPIs using a big-data architecture to generate a next load schedule for next cycle of fabrication.
  • KPIs current key performance indicators
  • the non-transitory computer-readable storage medium contains computer-executable instructions for, when executed by one or more processors, performing a semiconductor fabrication scheduling method.
  • the method includes creating a load scheduling data schema including facility data of product lots to be dispatched to a plurality of workstations; and generating a load schedule profile using a load-balancing model and based on the load scheduling data schema.
  • the load-balancing model includes one or more objective functions and there is at least one weight factor in an objective function.
  • the method also includes generating a current load schedule based on the load schedule profile; dispatching the product lots to the plurality of workstations using the current load schedule to complete fabrication of the product lots; obtaining a set of current key performance indicators (KPIs) of the completed fabrication of the product lots; and automatically adjusting the weight factors of the objective functions of the load-balancing model based on the current KPIs using a big-data architecture to generate a next load schedule for next cycle of fabrication.
  • KPIs current key performance indicators
  • FIG. 1 illustrates an exemplary semiconductor fabrication system consistent with the disclosed embodiments of the present invention
  • FIG. 2 illustrates an exemplary computing system consistent with the disclosed embodiments
  • FIG. 3 illustrates a functional/structural block diagram of an exemplary load scheduling system consistent with the disclosed embodiments
  • FIG. 4 illustrates an exemplary load scheduling process consistent with the disclosed embodiments
  • FIG. 5 illustrates an exemplary optimization process based on big-data to adjust the weight factors of the load-balancing model consistent with the disclosed embodiments
  • FIG. 6 illustrates an exemplary machine learning algorithm consistent with the disclosed embodiments.
  • FIG. 1 illustrates an exemplary semiconductor fabrication system 100 consistent with the disclosed embodiments of the present invention.
  • semiconductor fabrication system 100 may include a dispatching system 112, a plurality of work-stations 110-1, 110-2, ..., 110-J, a transfer bus 114, a database 116, a scheduling server 120, and a big-data computing platform 130, where J is a natural number indicating a total number of work-stations. Certain components may be omitted and other components may be added.
  • Dispatching system 112 may include any appropriate semiconductor wafer dispatching system that dispatches raw or processed wafers 118 (e.g., wafers, wafer lots, or product lots) to the pieces of manufacturing equipment (e.g., workstations) .
  • the dispatching system 112 may be a lot dispatcher.
  • the dispatching system 112 may be implemented on a computing system controlling a dispatching mechanism.
  • the plurality workstations 110-1, 110-2, ..., 110-J may represent a total J number pieces of manufacturing equipment, and each equipment may have one or more manufacturing stages. Each manufacturing equipment or workstation, and/or manufacturing stage may perform one or more processes in the fabrication of a wafer in the product.
  • One or more workstations may be considered as a manufacturing line for a specific integrated circuit (IC) fabrication, including front-end-of-lien (FEOL) processing and/or back-end-of-line (BEOL) processing.
  • IC integrated circuit
  • the workstations 110-1, 110-2, ..., 110-J may include any appropriate fabrication equipment, with a same type or different types.
  • the workstations may include deposition equipment, such as physical vapor deposition (PVD) workstations, chemical vapor deposition (CVD) workstations, electrochemical deposition (ECD) workstations, molecular beam epitaxy (MBE) workstations, and atomic layer deposition (ALD) workstations.
  • the workstations may also include wet or dry etch processing workstations, and/or chemical-mechanical planarization (CMP) workstations.
  • the workstations may also include optical lithography or UV lithography workstations. Further, the workstations may include doping workstations and/or annealing workstations. Other types of fabrication equipment may also be included.
  • a transfer bus 114 may be used to transfer product lots among the plurality work stations, the plurality manufacturing stages, and/or among different manufacturing lines.
  • the fabrication or manufacturing process for a particular product lot may start from the dispatching system 112 and follow a corresponding schedule to go through the workstation (s) or manufacturing stages to produce a partial or complete product (e.g., a fabricated wafer) .
  • Each workstation may collect and report a status of the product lot's progress or work-in-process (WIP) status describing the advancing of the product lot through the workstation. After the product lot goes through all the workstations, fabrication data such as quality and performance data may also be obtained and reported.
  • WIP work-in-process
  • the load of these workstations, and/or stages of a workstation may need to be maximized or optimally scheduled. That is, the workstations, and/or stages of a workstation may be scheduled using a load schedule generated to optimize the fabrication process.
  • the scheduling scheme may consider various objectives for scheduling the fabrication process, including a load balancing scheme.
  • the load balancing scheme may refer to efficiently utilize the workstations and/or stages of the workstations, i.e., balancing the load of the workstations.
  • the load balancing scheme may also refer to balancing other objectives for the workstation scheduling.
  • the term “scheduling” may be used exchangeably with the term “load balancing.
  • the load balancing may include various factors and constraints, and may depends on specific fabrication or applications. For example, those product lots that have a high priority for delivery must be processed expeditiously to meet committed schedules. At the same time, the utilization rate of the workstations and the quality and performance of the fabrication may remain desired.
  • the scheduling server 120 may control and schedule the load of the dispatching system 112 using a load balancing scheme based on a predetermined algorithm and specific data of the current cycle of fabrication.
  • the scheduling server 120 may include or communicate with a database 116 to obtain various data used in the control and scheduling of the dispatching system 112, such as facility data, product data, process data, etc.
  • Database 116 may a standalone database system or a distributed database system, and may include certain user interfaces for the user to query the database, to enter data, and/or to display data, etc.
  • the big data computing platform 130 may include one or more computer servers having hardware and software configured to perform data mining, data analysis, and artificial intelligence functions. Specifically, big data computing platform 130 may collect a large amount of fabrication data including the fabrication data of the dispatching system 112 and the various workstations under various load scenarios. The big data computing platform 130 may also obtain data from the database 116.
  • the scheduling server 120 may communicate with the big data computing platform 130 to obtain historical data, simulated data, and/or real-time data to adjust the load scheduling for the dispatching system 112.
  • the big data computing platform 130 and the scheduling server 120 are shown as two separated systems, the big data computing platform 130 and the scheduling server 120 may be integrated into one system or may be distributed in multiple systems.
  • the scheduling server 120 and/or the big data computing platform 130 may be implemented by any appropriate computing system.
  • Figure 2 illustrates an exemplary computing system 200 consistent with the disclosed embodiments.
  • computing system 200 may include a processor 202, a storage medium 204, a monitor 206, a communication module 208, a database 210, and peripherals 212. Certain devices may be omitted and other devices may be included.
  • Processor 202 may include any appropriate hardware processor or processors. Further, processor 202 can include multiple cores for multi-thread or parallel processing.
  • Storage medium 204 may include memory modules, such as ROM, RAM, flash memory modules, and mass storages, such as CD-ROM and hard disk, etc. Storage medium 204 may store computer programs for implementing various processes, when the computer programs are executed by processor 202.
  • peripherals 212 may include various sensors and other I/O devices, such as keyboard and mouse, and communication module 208 may include certain network interface devices for establishing connections through communication networks.
  • Database 210 may include one or more databases for storing certain data and for performing certain operations on the stored data, such as database searching.
  • the scheduling server 120 may obtain data from the database 116 and the big data computing platform 130 to perform load scheduling determination for the dispatching system 112, and may provide instructions to the dispatching system to start fabrication.
  • the instructions may include the listing of product lots being fabricated, the order of the execution of the processes necessary to produce the product lots, and/or the manufacturing equipment (workstations) used to perform the processes.
  • the status of each lot, the criticality factor of each piece of manufacturing equipment being employed in the process of each product lot, and the queue level of each piece of manufacturing equipment following the current piece of manufacturing equipment processing each of the product lots may also be obtained.
  • a measure of the priority necessary to balance the loading may be created such that the product lot is processed at an expeditious time for on-time delivery.
  • the scheduling server 120 may implement a load scheduling system for providing enhanced load balancing/scheduling for the dispatching system 112.
  • Figure 3 illustrates a functional/structural block diagram of an exemplary load scheduling system 300 consistent with the disclosed embodiments.
  • the load scheduling system 300 includes facility data 302, a data generation module 304, a scheduling module 306, a lot sequencing module 308, a load schedule 310, a data warehouse 312, a dispatching module 314, the big data computing platform 130, and a weight-factor file generation module 318.
  • Other components may also be included.
  • the facility data 302 may include any available data for the current fabrication, such as data of each available resource (e.g., workstation capacity, duration and order of stages and processes of the workstation) , data of the product (e.g., worked required for the product, priority, duration and order of product processing steps) , and data of the processes (e.g., standards, criteria, and performances) . Other types of data may also be included.
  • data of each available resource e.g., workstation capacity, duration and order of stages and processes of the workstation
  • data of the product e.g., worked required for the product, priority, duration and order of product processing steps
  • data of the processes e.g., standards, criteria, and performances
  • Data generation module 304 may include certain data processing functions (e.g., a query executor or a dispatcher) to convert the facility data and other types of data into one or more specific data schema or data tables to be used by the scheduling module 306.
  • the scheduling module 306 may include a load balancing model to generate a load schedule based on the data tables inputted by the data generation module 304.
  • the load balancing model may include one or more predetermined load balancing algorithms to generate a load schedule profile.
  • the load schedule profile may be information required to create a final load schedule 310 for the dispatcher, such as an initial load schedule, a load schedule to be verified, or any intermediate form of a final load schedule.
  • the lot sequencing module 308 may include a simulation model for simulating an actual fabrication process based on the load schedule profile such that the load schedule profile can be verified. After the verification, the sequencing module 308 may generate an actual load schedule 310 for the manufacturing equipment (e.g., the workstations) .
  • the load schedule may be a text file, or may be a file in a predetermined format, such as a spreadsheet file.
  • the load schedule 310 may be entered into the data warehouse 312.
  • the data warehouse 312 may refer to any database or computer server for hosting a large amount of data together with analysis tools to realize the data warehouse functionalities.
  • the data warehouse may accumulate all accepted load schedule files as well as all corresponding data, such as the fabrication scenarios for individual load schedules and other production data.
  • the fabrication scenarios may include information about the product lots, the workstations, the fabrication requirements, and production standards, etc.
  • the load schedule 310 may be provided to the dispatching module 314 (e.g., the dispatching system 120) for dispatching the wafer lots to the workstations to complete the fabrication processes.
  • the dispatching module 314 e.g., the dispatching system 120
  • the dispatching module 314 and/or the workstations can also provide fabrication data, such as production key performance indicators (KPIs) to the big data computing platform 130.
  • KPIs production key performance indicators
  • the load schedule 310 may also be provided to the big data computing platform 130.
  • the big data computing platform 130 may acquire all data needed to perform big data operations or machine learning operations, etc., and may provide results to the weight-factor file generation module 318.
  • the weight-factor file generation module 318 may include any appropriate analysis algorithm to determine various weight factors for the scheduling module 306 based on the results from the big data computing platform 130.
  • the scheduling system 300 shown in Figure 3 may perform a load scheduling process to create an optimized load schedule for completing the fabrication processes.
  • Figure 4 illustrates an exemplary load scheduling process 400 consistent with the disclosed embodiments.
  • the load scheduling data schema may include one or more data tables generated based on facility data and other types of data, such as lot predictions. These data tables may also be called as input data tables.
  • the facility data may be site-specific data, i.e., specific to a specific manufacturing site including the plurality of workstations.
  • the lot predictions may be data templates and data model specific data, such as solutions schema data.
  • a load schedule profile is generated based on the input data tables and a load-balancing model (S404) .
  • the input data tables may be provided to the scheduling module 306, the scheduling module 306 may first generate a plurality of data tables to be used by the load-balancing model.
  • the scheduling module 306 may generate one or more of: an objective function weight table, a parameter table, a resource table, a task table, a task capacity table, a task setup table, and/or a time bucket table, etc.
  • the load-balancing model may use the data in the generated data tables and generate a load schedule profile.
  • the load schedule profile may be in one or more output tables, such as a load balance table, and a lot assignment table, etc.
  • the load-balancing model may include any appropriate mathematical model configured to solve an assignment problem, i.e., finding an optimal assignment of tasks to resources based on related objective functions, such as balancing of work across the resources, with respect to the defined constraints (e.g., the limited capacity of the resources) .
  • the load-balancing model may be a mixed-integer programming (MIP) model.
  • MIP mixed-integer programming
  • the load-balancing model may be configured to model the assignment of lot-steps to the workstations, other applications that requires assignment of tasks may also be applicable.
  • the load-balancing model (or the scheduling model) may be an MIP model configured to find desired lot processing across scheduled tools (i.e., resources or workstations) .
  • the load-balancing model performs lot-to-tool assignments to meet specified scheduling objectives (such as minimizing tool setup changes, while meeting specified move targets) .
  • the load-balancing model may have one or more objective functions used to describe the relative cost of metrics that are optimized when producing a load scheduling setup.
  • the load-balancing model may include a load balance objective function.
  • the objective function is to minimize absolution deviation for each resource (workstation) :
  • AveWork refers to the average work for all resources, which may be calculated by dividing the sum of all work by the number of resources
  • Each resource has a limited capacity
  • the load-balancing model may include additional objective functions, which may be selected by a user and may include any one or more of: Full Bucket objective function, Constraints Move Target objective function, Process Early objective function, Setup Count objective function, Setup Time objective function, Per Station Early Processing objective function, Schedule Stability objective function, Reticle Count objective function, Non Preferred Resource objective function, Small Lots objective function, Distribute Tasks objective function, Seasoning objective function, Setup Change objective function, and Setup Limit objective function.
  • additional objective functions which may be selected by a user and may include any one or more of: Full Bucket objective function, Constraints Move Target objective function, Process Early objective function, Setup Count objective function, Setup Time objective function, Per Station Early Processing objective function, Schedule Stability objective function, Reticle Count objective function, Non Preferred Resource objective function, Small Lots objective function, Distribute Tasks objective function, Seasoning objective function, Setup Change objective function, and Setup Limit objective function.
  • a Load Balance objective function may refer to an objective functions that gives a cost to unbalanced loads, that is some resources starving for tasks while other resources are drowning in tasks.
  • a Full Bucket objective function may refer to an objective functions that gives a cost to time buckets (main time interval is split into a few time buckets) that are not nearly full. The objective is to make sure that time buckets are all nearly full, which helps reducing wasting resource time.
  • a Constraints Move Target objective function may refer to an objective functions that gives a cost to tasks that are not close to their move targets.
  • a Process Early objective function may refer to an objective functions that gives a cost to tasks scheduled later and this way makes sure that all tasks are scheduled as early as possible.
  • a Setup Time objective function may refer to an objective functions that gives a cost to the number of setups (setup times) used.
  • a Per Station Early Processing objective function may refer to an objective functions that gives a cost to tasks labelled as "process early" by the user when they are scheduled later.
  • a Schedule Stability objective function may refer to an objective functions that gives a cost to tasks that are scheduled in one optimization scenario but not in another. This reduces differences between scenarios and make the schedule more stable.
  • a Reticle Count objective function may refer to an objective functions that gives a cost to reticle use. It makes sure load on reticles are balanced, prevents overusing some and starving others.
  • a Non Preferred Resource objective function may refer to an objective functions that gives a cost to tasks that are not processed on a preferred resource.
  • a Small Lots objective function may refer to an objective functions that gives a cost when two or more tasks defined as small lots are processed on the same resource in the same time bucket.
  • a Distribute Tasks objective function may refer to an objective functions that gives a cost when two or more tasks that are assigned to a certain bucket are processed on the same resource.
  • a Seasoning objective function may refer to an objective functions that gives a cost when two or more tasks requiring different PPIDs (process program ID) are assigned to the same resource.
  • a Setup Change objective function may refer to an objective functions that gives a cost when two tasks requiring different setups are assigned to the same resource.
  • a Setup Limit objective function may refer to an objective functions that gives a cost when too many setups of a certain type are assigned to a resource within a specified number of buckets beginning with the first bucket.
  • objective functions may also be included.
  • the user may select a specific set of objective functions for a current fabrication cycle (i.e., from load scheduling to complete of the fabrication process) .
  • a load schedule may be generated based on the load schedule profile (e.g., the output data tables) (S406) .
  • the lot sequencing module 308 may use an AutoSched AP/Fusion engine based on the output data tables from the load-balancing model to generate the load schedule for the plurality of workstations.
  • the load schedule may include a Lot Assignments table, which assigns specific lots to scheduled workstations for processing.
  • the dispatcher may dispatch the product lots to the plurality of workstations based on the load schedule and to start the fabrication production (S408) . That is, the dispatcher may use the Lot Assignments table to load the production lots to scheduled workstations and to complete the fabrication processes.
  • the load schedule may also be stored in the data warehouse for later retrieval.
  • the load-balancing model (or the scheduling model) may be optimized automatically for next fabrication cycle using a big-data architecture (S410) .
  • KPIs production key performance indicators
  • the load-balancing model may be automatically optimized for next fabrication cycle.
  • the criteria may include a set of target KPIs for the current fabrication (e.g. target cycle time or wafer moves) .
  • the determination may be performed based on a comparison between the current KPIs and target KPIs stored on the server.
  • the determination may also be performed based on a comparison between the current KPIs and the historical KPIs.
  • the server may store various target KPIs, and each set of KPIs may correspond to a specific fabrication scenario, so that the current KPIs can be compared with the target KPIs under a same or similar fabrication scenarios.
  • the load-balancing model may be adjusted to perform the optimization. That is, the optimization may include adjusting parameters of the load-balancing model to improve the load schedule and the fabrication performance.
  • the load-balancing model may include one or more objective functions. For each objective function, there may be one or more auxiliary variables (or weight factors) that can be adjusted. That is, after the current cycle of fabrication and before the next cycle of fabrication, the weight factors of certain or all objective functions of the load-balancing model may be adjusted automatically based on a big-data mechanism.
  • Figure 5 illustrates an exemplary optimization process based on big-data to adjust the weight factors of the load-balancing model consistent with the disclosed embodiments.
  • the big-data computing platform 130 may acquire various data for optimization of the load-balancing model (S502) .
  • a big-data architecture is used in the sense that the acquired data is broad and in a large amount, as the acquired data may be both historical and current, from current fabrication site and from other fabrication site and under various fabrication scenarios. Thus, such big-data architecture may produce more accurate results for adjusting the load-balancing model.
  • the big-data computing platform 130 may acquire data about the load schedule, and the production KPIs, etc.
  • the production KPIs may indicate the current status of the actual fabrication environment and production performance.
  • the KPIs may include one or more of: load balancing performance, bullet lot cycle time, hot lot cycle Time, actual cycle time, preferred tools, meet move targets, lot step due date, schedule stability, and reduce setup changes, etc.
  • Load balancing performance may refer to a KPI that measures total work time or percentage for every resource.
  • Hot lot cycle Time may refer to a KPI that measures cycle time for bullet lots, lots that receive preferential treatment in the fab and that are processed in front of the queue at process tools.
  • Bullet lot cycle time may refer to a KPI that measures cycle time for bullet lots, lots that receive even more preferential treatment in the fab and that are processed in at the very front of the queue at process tools.
  • Actual cycle time may refer to a KPI that measures cycle time, the number of days to complete a lot.
  • Preferred tools may refer to a KPI that measures the scheduling on preferred tools, a way to indicate whether preferential assignment can be increased or should be decreased in the fab.
  • Meet move targets may refer to a KPI that measures the daily move targets (production capacity) achieved in the fab production line.
  • Lot step due date may refer to a KPI that measures whether processing of lots meet due date at every step in the fab.
  • Schedule stability may refer to a KPI that measures the stability of the schedule produced by the optimizer in the production line. The differences between the theoretical schedule and in practice is monitored.
  • Setup changes may refer to a KPI that measures the setup changes performed at process tools to apply the schedule.
  • the big-data computing platform 130 may acquire historical data as well as the real-time date (i.e., data for the current fabrication cycle) .
  • the big-data computing platform 130 may acquire load schedules used in the past and its corresponding KPIs under the same fabrication scenarios.
  • the big-data computing platform 130 may calculate scores of the load schedules in the acquired data for optimization (S504) .
  • the load-balancing model may be optimized to maximize the score of the load schedule generated by the load-balancing model.
  • a score can be calculated for each load schedule.
  • the score may reflect performance of the fabrication based on the load schedule. For example, a high score may reflect desired performance of the current fabrication, i.e., desired KPIs.
  • the score may be derived by any appropriate methods.
  • the score of the load schedule may be calculated based on the applicable objective functions of the load schedule.
  • the score may be calculated based on the values of the objective functions and corresponding auxiliary variables or weight factors of the objective functions. Specifically, a value of each objective function of the load-balancing model may be determined and, as each objective function may have a weight factor, the score S of the load schedule determined by the load-balancing model may be calculated as:
  • the aim of the scheduling optimization is to find the best combination of weight factors to realize a highest score.
  • the set of weight factors may be determined in advance based on a variable list or may be selected by the user.
  • the big-data computing platform 130 may use a predetermined algorithm to optimize the load-balancing model (S506) .
  • the big-data computing platform 130 may apply a genetic algorithm, a Monte Carlo algorithm, a machine learning algorithm, or other similar optimization algorithms to perform the optimization. That is, the big-data computing platform 130 may use a genetic algorithm or a Monte Carlo algorithm to determine a best set of weight factors for the load-balancing model to achieve a highest score for the load schedule generated by the load-balancing model based on current constraints on the load-balancing model.
  • the big-data computing platform 130 may use a machine learning algorithm to perform the optimization.
  • Figure 6 illustrates an exemplary machine learning algorithm (e.g., a deep learning neural-network model) 600 consistent with the disclosed embodiments.
  • the deep-learning model 600 may include an input layer 604, middle layers 606 and 608 (hidden layers) , and an output layer 610, etc.
  • Input 602 are provided to the input layer 604, and output 612 are provided by the output layer 610.
  • Each layer may include one or more neural-network node. The number of the neural-network layers are used for illustrative purposes, any number of the neural-network layers may be used.
  • the neural-network deep-learning model 600 may be used as an example for illustration, other architectures may also be used similarly.
  • the big-data computing platform 130 may first train the deep-learning model 600 using relationships between the production KPIs and scores of load schedules.
  • previous load schedules and corresponding values of a set of production KPIs may be obtained.
  • the previous load schedules may be from various fabrication scenarios and workstations, or may be from a same fabrication scenario and same workstations but with different product lots.
  • the objective functions and objective function weight factors may be obtained or determined, and the score and/or weight factors of each load schedule may be calculated.
  • the deep-learning model 600 may be trained based on the historical production KPIs and corresponding scores and/or values of weight factors. That is, the production KPIs are provided as input 602 of deep-learning model 600, and the scores and/or values of weight factors are provided as output 612 of deep-learning model 600. That is, in certain embodiment, the output of the deep-learning model 600 is the score, and the score may be used to derive corresponding values of the weight factors. In certain other embodiment, the values of the weight factors may be the output of the deep-learning model 600 directly, and score can be calculated based on the values of the weight factors. In certain other embodiments, both the scores and the values of the weight factors may be the output of the deep-learning model 600.
  • the target values of current set of production KPIs may be provided as input to the deep-learning model 600, and the score of the load schedule and the values of the weight factors of the load-balancing model may then be obtained based on the output of the deep-learning model 600.
  • the score of the current load schedule may be calculated and may be compared with the obtained score from the deep-learning model 600. If it is determined that the obtained score is greater than the score of the current load schedule, the obtained values of the weight factors from the deep-learning model 600 may be used as the optimized values of the weight factors.
  • the adjusted load-balancing model is used to generate an improved load schedule and the next fabrication cycle may be started based on the improved load schedule (S412) . Further, the production KPIs of the next fabrication cycle is obtained and whether the production KPIs meet the preset criteria is determined. The above optimization cycle can then be repeated automatically until the production KPIs of the fabrication cycle meet the preset criteria (S414) .
  • the weight adjustment is not based on human trials and understanding of the objective functions, and manually adjusting the weight factors of the load-balancing model can be avoided. Further, any misunderstanding or miscommunication between human operators and any mistake may be the human operator in the operation of the manufacturing equipment can also be avoided.
  • a typical cycle for weight adjustment by human trials takes approximately a few days to a week, which is time consuming and affects the production line KPI (e.g. cycle time and wafer moves) during this adjustment period, manufacturing cost, efficiency, and effectiveness of the IC fabrication can be greatly improved.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Business, Economics & Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Human Resources & Organizations (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Manufacturing & Machinery (AREA)
  • Automation & Control Theory (AREA)
  • Quality & Reliability (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Computational Linguistics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Physiology (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computer Hardware Design (AREA)
  • Power Engineering (AREA)
  • Molecular Biology (AREA)
  • Development Economics (AREA)

Abstract

Un procédé de planification de fabrication de semi-conducteurs consiste à créer un schéma de données de planification de charge comprenant des données d'installation de lots de produits à envoyer à une pluralité de postes de travail ; à générer un profil de planification de charge à l'aide d'un modèle d'équilibrage de charge et sur la base du schéma de données de planification de charge, le modèle d'équilibrage de charge comprenant une ou plusieurs fonctions objectives et au moins un facteur de pondération étant présent dans une fonction objective ; à générer une planification de charge courante sur la base du profil de planification de charge ; à distribuer des lots de produits à la pluralité de stations de travail à l'aide de la planification de charge courante pour achever la fabrication des lots de produits ; à obtenir un ensemble d'indicateurs de performances clés actuels (KPI) de la fabrication achevée des lots de produits ; et à ajuster automatiquement les facteurs de pondération des fonctions objectives du modèle d'équilibrage de charge sur la base des KPI actuels à l'aide d'une architecture de mégadonnées pour générer une planification de charge suivante pour le prochain cycle de fabrication.
PCT/CN2020/076111 2020-02-21 2020-02-21 Procédé et système de planification de fabrication de semi-conducteurs WO2021163986A1 (fr)

Priority Applications (4)

Application Number Priority Date Filing Date Title
PCT/CN2020/076111 WO2021163986A1 (fr) 2020-02-21 2020-02-21 Procédé et système de planification de fabrication de semi-conducteurs
CN202080000366.2A CN111356988B (zh) 2020-02-21 2020-02-21 用于调度半导体加工的方法和系统
TW109110944A TWI788650B (zh) 2020-02-21 2020-03-31 用於半導體加工調度的方法、半導體加工調度系統和非暫時性電腦可讀儲存介質
US16/905,194 US11385627B2 (en) 2020-02-21 2020-06-18 Method and system for scheduling semiconductor fabrication

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/076111 WO2021163986A1 (fr) 2020-02-21 2020-02-21 Procédé et système de planification de fabrication de semi-conducteurs

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/905,194 Continuation US11385627B2 (en) 2020-02-21 2020-06-18 Method and system for scheduling semiconductor fabrication

Publications (1)

Publication Number Publication Date
WO2021163986A1 true WO2021163986A1 (fr) 2021-08-26

Family

ID=71197887

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/076111 WO2021163986A1 (fr) 2020-02-21 2020-02-21 Procédé et système de planification de fabrication de semi-conducteurs

Country Status (4)

Country Link
US (1) US11385627B2 (fr)
CN (1) CN111356988B (fr)
TW (1) TWI788650B (fr)
WO (1) WO2021163986A1 (fr)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2022523870A (ja) * 2019-03-29 2022-04-26 ラム リサーチ コーポレーション 基板処理システムのためのモデルベースのスケジュール設定
CN113361833B (zh) * 2020-03-02 2022-05-24 联芯集成电路制造(厦门)有限公司 化学机械抛光系统以及相关的派工管理方法
US11901204B2 (en) * 2020-05-22 2024-02-13 Applied Materials, Inc. Predictive wafer scheduling for multi-chamber semiconductor equipment
CN112101717B (zh) * 2020-08-07 2024-04-19 青岛奥利普奇智智能工业技术有限公司 调整产线的生产任务的方法、装置、设备和存储介质
EP3992878B1 (fr) * 2020-09-11 2023-08-09 Changxin Memory Technologies, Inc. Procédé et appareil de traitement de tâche d'inspection de tranche, système et support de stockage
WO2023079562A1 (fr) * 2021-11-08 2023-05-11 Augury Systems Ltd. Mise en œuvre de paramètres de fonctionnement dans des processus industriels

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110190917A1 (en) * 2010-01-29 2011-08-04 James Moyne Method and apparatus for developing, improving and verifying virtual metrology models in a manufacturing system
CN103217960A (zh) * 2013-04-08 2013-07-24 同济大学 一种半导体生产线动态调度策略自动选择方法
CN103439886A (zh) * 2013-07-26 2013-12-11 同济大学 一种半导体生产线自适应动态调度装置
CN105045243A (zh) * 2015-08-05 2015-11-11 同济大学 一种半导体生产线动态调度装置
CN109085803A (zh) * 2018-07-06 2018-12-25 昆明理工大学 一种dram内存封装过程的优化调度方法
US20190354089A1 (en) * 2018-05-18 2019-11-21 Taiwan Semiconductor Manufacturing Co., Ltd. Method, system and non-transitory computer-readable medium for improving cycle time

Family Cites Families (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6714001B2 (en) * 2001-11-28 2004-03-30 Winbond Electronics Corporation Dispatching method of manufacturing integrated circuit
US7426420B2 (en) * 2003-09-15 2008-09-16 International Business Machines Corporation System for dispatching semiconductors lots
CN102446786B (zh) * 2011-11-28 2013-12-04 上海华力微电子有限公司 一种半导体制程中的设备监控方法
CN104217978B (zh) * 2013-06-05 2017-05-17 中芯国际集成电路制造(上海)有限公司 半导体批次产品的处理系统和方法
CN103310285A (zh) * 2013-06-17 2013-09-18 同济大学 可用于半导体生产线动态调度的性能预测方法
CN104423331B (zh) * 2013-08-20 2017-06-16 中芯国际集成电路制造(上海)有限公司 半导体集成电路生产中晶圆制造调度方法及调度系统
CN104576441B (zh) * 2013-10-29 2017-07-07 中国科学院沈阳自动化研究所 基于作业区域的半导体封装线lot多规则调度方法
US9748088B2 (en) * 2013-12-09 2017-08-29 Globalfoundries Inc. Method, storage medium and system for controlling the processing of lots of workpieces
CN105182916B (zh) * 2014-06-18 2019-06-18 上海华力微电子有限公司 半导体生产约束管控方法
US10268186B2 (en) * 2014-09-25 2019-04-23 Taiwan Semiconductor Manufacturing Co., Ltd. System and method for dispatching lot
CN105843180A (zh) * 2015-01-15 2016-08-10 中芯国际集成电路制造(上海)有限公司 基于半导体加工设备生产配方的派货系统及方法
CN106444649B (zh) * 2016-07-05 2019-01-11 同济大学 一种半导体生产线闭环调度控制方法
US10636007B2 (en) * 2017-05-15 2020-04-28 Tata Consultancy Services Limited Method and system for data-based optimization of performance indicators in process and manufacturing industries
CN109146212B (zh) * 2017-06-16 2022-03-25 佛山科学技术学院 众包系统中的大规模同构任务分配方法
KR102491068B1 (ko) * 2017-11-17 2023-01-19 에스케이하이닉스 주식회사 메모리 장치에 대한 태스크들을 스케줄링하는 반도체 장치 및 이를 포함하는 시스템
US11042148B2 (en) * 2017-11-23 2021-06-22 Taiwan Semiconductor Manufacturing Company, Ltd. System and method for scheduling semiconductor lot to fabrication tool
CN108171372B (zh) * 2017-12-25 2022-06-14 上海大学 一种存在准备时间的多品种生产分批调度方法
US11100437B2 (en) * 2018-05-10 2021-08-24 Applied Materials, Inc. Method for improving semiconductor back-end factories
CN108829052A (zh) * 2018-05-31 2018-11-16 同济大学 一种半导体生产线cps环境下的调度优化方法及装置
CN108873835A (zh) * 2018-06-12 2018-11-23 昆明理工大学 一种半导体集成电路制造中光刻工艺的优化调度方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110190917A1 (en) * 2010-01-29 2011-08-04 James Moyne Method and apparatus for developing, improving and verifying virtual metrology models in a manufacturing system
CN103217960A (zh) * 2013-04-08 2013-07-24 同济大学 一种半导体生产线动态调度策略自动选择方法
CN103439886A (zh) * 2013-07-26 2013-12-11 同济大学 一种半导体生产线自适应动态调度装置
CN105045243A (zh) * 2015-08-05 2015-11-11 同济大学 一种半导体生产线动态调度装置
US20190354089A1 (en) * 2018-05-18 2019-11-21 Taiwan Semiconductor Manufacturing Co., Ltd. Method, system and non-transitory computer-readable medium for improving cycle time
CN109085803A (zh) * 2018-07-06 2018-12-25 昆明理工大学 一种dram内存封装过程的优化调度方法

Also Published As

Publication number Publication date
TWI788650B (zh) 2023-01-01
US20210263505A1 (en) 2021-08-26
TW202133097A (zh) 2021-09-01
US11385627B2 (en) 2022-07-12
CN111356988B (zh) 2021-03-12
CN111356988A (zh) 2020-06-30

Similar Documents

Publication Publication Date Title
US11385627B2 (en) Method and system for scheduling semiconductor fabrication
EP0679972B1 (fr) Méthode et appareil de commande et de coordination de machines de traitement
US7305276B2 (en) Method, system, and computer program product for controlling the flow of material in a manufacturing facility using an extended zone of control
CN104915251B (zh) 任务调度方法及装置
CN108053047A (zh) 云生产资源调度方法、装置和系统
CN113705866B (zh) 基于资源受限项目调度问题模型的排程优化方法和系统
JP2000077289A (ja) 製造予測管理システム
CN116529741A (zh) 基于仿真的闭环aps调度优化方法、系统及存储介质
Kutin et al. Simulation modeling of assembly processes in digital manufacturing
Ištoković et al. Simulation framework for determining the order and size of the product batches in the flow shop: A case study
Singh et al. Experimental investigation for performance assessment of scheduling policies in semiconductor wafer fabrication—a simulation approach
Thenarasu et al. Development and selection of hybrid dispatching rule for dynamic job shop scheduling using multi-criteria decision making analysis (MCDMA)
Laipple et al. Generic data model for semiconductor manufacturing supply chains
CN116540659A (zh) 一种大型复杂产品车间调度方法、系统、设备及介质
CN115660261B (zh) 生产订单信息处理方法、计算机装置和存储介质
US20050250226A1 (en) Method for generating work-in-process schedules
Talapatra et al. Overall efficiency improvement of a production line by using Yamazumi chart: a case study
CN114418418B (zh) 基于工序换算比的工单信息流转方法、系统、设备及介质
US20090313058A1 (en) Tool, method and system for project management
CN115545329A (zh) 一种适用于服装制造产线的一体化动态排程方法
Hung et al. Sensitivity search for the rescheduling of semiconductor photolithography operations
KORUCA et al. A priority rule based production scheduling module on faborg-sim simulation tool
Lamghari-Idrissi et al. Influence of spare parts service measures on the performance of front-end wafer production process
Fordyce et al. A dynamically generated rapid response capacity planning model for semiconductor fabrication facilities
Sugimura et al. A study on integrated process planning and scheduling system for holonic manufacturing

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20919776

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20919776

Country of ref document: EP

Kind code of ref document: A1